The present invention generally pertains to user interaction with a computing device. More specifically, the present invention pertains to user interactions with a centralized interface in the context of speech recognition.
Much attention, lately, has been directed towards the improvement of speech recognition technology. One such challenge in speech recognition technology lies in user interaction with a computing device using voice commands. Frequently, a voice command needs further clarification before the computing device can execute such a command.
A computing device can often interpret a voice command in multiple ways. In one aspect, a computing device may not understand which application the voice command is directed towards. For example, a voice command can include terms that are used across a plurality of application modules. In another aspect, an application may not understand what the user would like to execute because the voice command contains ambiguous information. For example, a voice command can include “play the Beatles” when there are multiple Beatles albums that the application could play. In still another example, the voice command can contain a misrecognition error. For example, the user may recite the command “insert ice cream” while the speech recognition system recognizes the command as “insert I scream”.
Making guesses at what the user meant or taking action without consulting the user can lead to frustration for a user. There is a need to provide the user with the tools to control the clarification of voice commands and clarification of misrecognition errors without causing undue frustration and sacrificing speech recognition efficiency.